from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer
import torch
from PIL import Image
import streamlit as st

st.title('Image Caption Generator for Listed - 1')
uploaded_file = st.file_uploader('Choose an image', type=['jpg', 'jpeg', 'png'])
num_return_sequences = st.number_input('Number of captions to generate', min_value=1, max_value=10, value=1)


model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
feature_extractor = ViTImageProcessor.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning")

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)



max_length = 16
num_beams = 4

gen_kwargs = {"max_length": max_length, "num_beams": num_beams, "num_return_sequences": num_return_sequences}
def predict_step(image_paths):
  images = []
  for image_path in image_paths:
    i_image = Image.open(image_path)
    if i_image.mode != "RGB":
      i_image = i_image.convert(mode="RGB")

    images.append(i_image)

  pixel_values = feature_extractor(images=images, return_tensors="pt").pixel_values
  pixel_values = pixel_values.to(device)

  output_ids = model.generate(pixel_values, **gen_kwargs)

  preds = tokenizer.batch_decode(output_ids, skip_special_tokens=True)
  preds = [pred.strip() for pred in preds]
  return preds


if uploaded_file is None:
    st.write("Please upload an image.")
else:
    captions = predict_step([uploaded_file])
    st.write(captions)